So production.log is 4G....
If I'm going to setup logrotate, do I have to have wagn restart postrotate?
I'm just looking at the config file in /etc/logrotate.d for apache2. Everything else in there seems like it would be fine, but the postrotate and prerotate things seem apache-specific, and if I need to tell wagn to reload, well, I don't know how to do that.
Can I just drop those parts and be ok?
/srv/www.teenwolfwiki.com/log/*.log {
weekly
missingok
rotate 52
compress
delaycompress
notifempty
create 640 root adm
sharedscripts
postrotate
/etc/init.d/apache2 reload > /dev/null
endscript
prerotate
if [ -d /etc/logrotate.d/httpd-prerotate ]; then \
run-parts /etc/logrotate.d/httpd-prerotate; \
fi; \
endscript
}
I don't know anything about logrotate. A quick glance made it look like it's run as a cron job, though, so my guess is you don't have to do anything.
Well, I guess the question is this: is wagn going to throw a fit if production.log suddenly changes while it's running?
FWIW
/srv/www.teenwolfwiki.com/log/*.log {
weekly
missingok
rotate 52
compress
delaycompress
notifempty
sharedscripts
postrotate
/etc/init.d/apache2 reload > /dev/null
endscript
prerotate
if [ -d /etc/logrotate.d/httpd-prerotate ]; then \
run-parts /etc/logrotate.d/httpd-prerotate; \
fi; \
endscript
}
seems to work quite nicely. My 4G log file is now a 340MB log file and it's already done a second rotation. I don't actually know if the reloading apache part is necessary, but it doesn't seem to be hurting anything.